Polyglutamine (polyQ) diseases are a family of slow-progressing neurodegenerative disorders caused by CAG triplet repeat expansions within the coding regions of distinct and unrelated genes. Spinal and bulbar muscular atrophy (SBMA) is one such disease that is characterized by a loss of brain stem and spinal cord motor neurons, and of the associated innervated muscles. This toxicity to the neuromuscular system is caused by the expansion of a CAG repeat-encoded polyQ segment within the androgen receptor protein, a transcription factor that is activated by its cognate ligands, testosterone and dihydrotestosterone. Although the molecular events that mediate expanded-polyQ-dependent toxicity remain largely obscure, such long polyQ tracts are thought to cause cellular dysfunction and ultimately cell death by dysregulating protein-protein interactions that sustain normal cellular function. Therefore, to understand this dysregulation, we have employed a quantitative proteomics approach to identify changes in the AR protein interaction network caused by polyQ expansion. In five independent experiments, one of the top hits identified was the ubiquitin-specific protease USP7, which we have validated as a preferential interactor with polyQ-expanded AR. We also observed that USP7 interacts with polyQ- expanded AR in SBMA transgenic mice. Moreover, while partial knockdown of USP7 reduced polyQ-expanded AR aggregation and cytotoxicity, overexpression of wild-type, but not catalytically inactive, USP7 resulted in a dramatic increase in polyQ-expanded AR aggregation as well as cytotoxicity. These findings support the idea that the deubiquitinase activity of USP7 plays a role in polyQ-expanded AR toxicity and that inhibiting USP7 activity may be a viable therapeutic strategy for the treatment of SBMA.